# AdaTask: Adaptive Multitask Online Learning

arXiv: https://arxiv.org/abs/2205.15802 · HTML: https://mlsanity.com/paper/2205.15802/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 38 references that are in the map.
- First posted on arXiv 31 May 2022.

## Builds on (most cited first, 13 of 13)

- [A Survey on Multi-Task Learning](https://mlsanity.com/paper/1707.08114/) (2017)
- [Introduction to Online Convex Optimization](https://mlsanity.com/paper/1909.05207/) (2019)
- [Online Learning: A Modern Introduction Using Convex Optimization](https://mlsanity.com/paper/1912.13213/) (2019)
- [Black-Box Reductions for Parameter-free Online Learning in Banach Spaces](https://mlsanity.com/paper/1802.06293/) (2018)
- [Provable Guarantees for Gradient-Based Meta-Learning](https://mlsanity.com/paper/1902.10644/) (2019)
- [A Principled Approach for Learning Task Similarity in Multitask Learning](https://mlsanity.com/paper/1903.09109/) (2019)
- [Regret Bounds for Lifelong Learning](https://mlsanity.com/paper/1610.08628/) (2016)
- [Multi-Task Learning with Labeled and Unlabeled Tasks](https://mlsanity.com/paper/1602.06518/) (2016)
- [Convex Learning of Multiple Tasks and their Structure](https://mlsanity.com/paper/1504.03101/) (2015)
- [Trace norm regularization for multi-task learning with scarce data](https://mlsanity.com/paper/2202.06742/) (2022)
- [Multitask Online Mirror Descent](https://mlsanity.com/paper/2106.02393/) (2021)
- [Nonparametric Online Regression while Learning the Metric](https://mlsanity.com/paper/1705.07853/) (2017)
- [Updating Singular Value Decomposition for Rank One Matrix Perturbation](https://mlsanity.com/paper/1707.08369/) (2017)

## Built on by (most cited first, 1 of 1)

- [Feedback Control for Multi-Objective Graph Self-Supervision](https://mlsanity.com/paper/2602.05036/) (2026)

## Abstract (arXiv)

We introduce and analyze AdaTask, a multitask online learning algorithm that adapts to the unknown structure of the tasks. When the N tasks are stochastically activated, we show that the regret of AdaTask is better, by a factor that can be as large as √N, than the regret achieved by running N independent algorithms, one for each task. AdaTask can be seen as a comparator-adaptive version of Follow-the-Regularized-Leader with a Mahalanobis norm potential. Through a variational formulation of this potential, our analysis reveals how AdaTask jointly learns the tasks and their structure. Experiments supporting our findings are presented.